Dr. Sonali Jhanjee

Psychiatry-Thesis                                                                     ISSN 1930-6741            

                                                                              SCREENING AN URBAN POPULATION

FOR

PSYCHIATRIC MORBIDITY

         A DISSERTATION SUBMITTED BY                                           DR. SONALI ARORA

FOR THE DEGREE OF M.D. BRANCH XI                (PSYCHIATRY)

EXAMINATION OF POONA UNIVERSITY

 MAY, 1998

                                                             NAME OF SUPERVISOR – DR. N. R. PANDE.

                                                                                                INDEX

N O.                TOPIC                                                PAGE NO.

1.               INTRODUCTION                                         1.

2.               AIMS AND OBJECTIVES                           4.

3.               REVIEW OF LITERATURE                               5.

4.               MATERIALS AND METHODS                  19.

5.               RESULTS                                                      28.

6.               DISCUSSSION                                                      53.

7.               CONCLUSION                                             72.

8.     BIBLIOGRAPHY                                         77.

Community mental health is concerned with the prevention and treatment of mental disorder and with the rehabilitation of former psychiatric patients through the use of organized community programmers [Kaplan, 1994] The main focus of this movement has been shifting the locus of service provision away from the institution and towards the community [Mechanic, 1992].In the Indian situation, this movement has been initiated against the backdrop of growing needs of universal coverage of mental health services and has included a rapid transition from mental hospitals to General Hospitals Psychiatric Units (GHPUs) to community care [Srinivas Murthy, 1992].

The culmination of all these developments has been the National Mental Health Program (1982) whose objective are to ensure availability and accessibility of minimum mental health care for all, particularly the most vulnerable and underprivileged sections of the society.

In order to attain the above objectives, planning of psychiatric facilities is essential and that would be very much facilitated by epidemiological surveys which would give an idea of mental health problems in the community [Verghese, 1973].

In Pune, the Regional Mental Health Hospital, Yerwada, and the Department of Psychiatry, Sassoon General Hospitals (SGH) are providing Institution based services for manifest psychiatric morbidity.

Although surveys in the community have often been handicapped by the lack of valid and reliable methods of case identification, there is a wide spread consensus that statistics derived from patients under treatment give a incomplete estimate of the total psychiatric morbidity [Goldberg, 1970]. To overcome this bias a measure of untreated morbidity is required in the community and a very appropriate research strategy for such measurement is the two stage screening survey [Varquez Barquero, 1978].

Whereas estimates of severe psychiatric morbidity lie in the range of 1-2%, minor psychiatric morbidity lies within the range of 8-10% [Nandi et al, 1975]. Thus, as a case example, in the city of Pune itself one would expect around 1.7 lakh people to be suffering from minor psychiatric illness warranting specialist attention [Population 17, 35, 171, 1991 census].

In view of the size of the problem the parent institute is undertaking a project to establish mental health services for urban population. As part of this project, a feasibility survey is being carried out and it is envisaged that this survey would help us find out the nature of psychosocial services which the urban population needs [treatment of severe psychiatric morbidity would be carried out in the same way as currently practised].

One component of this survey comprises of screening the urban population for psychiatric morbidity with a view to validate the screening tool  so that it may help the parent institute for subsequent assessment in future.

 AIMS AND OBJECTIVES

 AIMS:

The study aims to assess psychiatric morbidity in an urban community using standard two-stage methodology with a view to

1.               Validate the screening tool.

2.               To use this methodology for subsequent assessment of pattern and prevalence of psychiatric morbidity in the urban population by the parent institute.

 OBJECTIVES:

1.               To document socio-demographic information of the study population.

2.               To screen the study population for presence of psychiatric morbidity using standard screening tool.

3.               To carry out validation of the screening test to establish cut-off scores.

4.               To carry out descriptive analysis (stratified) on the collected demographic and morbidity data.

REVIEW OF LITERATURE

In the field of mental health care, since the world war II, supported by development of specific therapeutic measures, care of the mentally ill in non-institutional settings has been the focus of attention. This approach has been referred to as the Community Mental Health (CMH) movement [Srinivas Murthy, 1992].

This was done to reduce the census of psychiatric hospitals which were located at a distance from the homes and families of the patients [Mechanic, 1992].

Within this context of de-institutionalisation, planning of public mental health service system became essential and for that epidemiological studies which would throw light on the prevalence and distribution of mental illness in the community became imperative.

Mental disorder epidemiology is the quantitative study of distribution and causes of mental disorder in the human population [Kaplan, 1995].

1.0          Methodological issues:

The search to determine the extent of mental health is complicated by 2 problems:

1)     detection of a case, and

2)     its diagnosis [Chakorborty, 1990]

Although the focus has shifted to community care, case definition criteria have lagged behind remaining rooted in concepts derived from hospital patients [Sen et al, 1987].

As Dohrenwend (1982) points out, “it is as if in a study of the community prevalence of diabetes, diabetes were defined by the presence of diabetic retinopathy”.

Hence owing to the different definition and sampling procedures any literature review of the prevalence of mental disorder should be made with some caution [Reichenheim, 1991].

Early efforts to carry out research in this direction used mental hospital and admission rates as a reference point [Carswell J, 1924].

It took the coincidental phenomenon of highly stable communities and diligent painstaking psychiatric researchers to yield the first really meaningful community prevalence studies [Sartorius, 1983].

2.0          First generation studies:

Although long-standing diagnostic confusion renders comparison difficult, Dohrenwend (1982) has called the older pre-world war II studies – the First generation studies.

These studies lacked standardised tools and used medical records and key informants to ascertain information. They also treated psychiatric disorders as a unified concept and hence could not generate rates for different diagnostic groups [Newton, 1962].

Beyond their unique contribution to the knowledge of psychosocial determinants of mental illness, these studies opened crucial questions on validity and feasibility of the application of mass diagnostic procedures in the field and posed to clinical psychiatry the challenge of developing more precise diagnostic assessment tools [Mezzich, 1994].

3.0 Second generation studies:

After Worldwar-II, the high prevalence of mental illness prompted studies in the general populations. These were called Second Generation studies by Dohrenwend (1982).

These studies had certain similarities:-

1.               They used impeccable method of sampling paying attention to the repesentativeness of the sample selected for the interviews and to the rates of  response.

2.               They decided against using existing psychiatric nosology out of an awareness of diagnostic unrealiability and substituted measures of overall mental impairment for traditional diagnostic categories. This made it easier and more economical to execute surveys.

3.               Highly trained psychiatrists were no longer required to make diagnostic judgements.

4.               Unitary concept of mental illness was embraced and hence they measured undifferentiated severity of psychiatric impairment.

5.               This golden era of social epidemiology was guided by the paradigm of social psychiatry [Weissmann, 1986].

Four landmark studies are representative of this era:

1.               Stirling Country Study [Leighton et al, 1963].

2.               Baltimore morbidity survey [Pasamanick, 1956].

3.               Mid-town Manhattan study [Srole L, 1962].

4.               Longitudinal study in New Haven [Weissmann, 1978].

In the first three studies, information was recorded on structured interview protocols by non-clinical interviewers and subsequently rated by a psychiatrist. Providing an important transition was the NewHaven study which obtained the diagnosis according to predefined operationalised criteria.

                                All studies surveyed urban population except the Stirling study (rural population). The sample size ranged from 511 in New Haven to 1,660 in Manhattan. These samples were considered to be sufficiently large to obtain an adequate number and distribution of cases at that time.

The Stirling study used American Psychiatric Association (APA) – Diagnostic and Statistical Manual (DSM) – I criteria for diagnosis and was testing the association of socio-cultural disintegration indexes while the Baltimore study used the same criteria but just to determine prevalence but made no attempt to assess the reliability and validity of their assessment.

The Midtown Manhattan study also testing for association of sociocultural variables with mental illness, opted for a unified concept of mental disorder and psychological impairment was rated on a gradient scale. No diagnosis was made. The study was severely criticised both on methodology and findings as they were not comparable to other studies. Nevertheless this study remains a landmark and is one of the few studies that have been conducted on densely populated urban and industrial areas. The New Haven study which was a longitudinal study introduced Schedule for Affective Disorders and Schizophrenia (SADS-L) to make a specific diagnosis.

 Results:

These results emphasised the lack of uniformity and comparability that existed on data on prevalence of surveys which would provide uniform basic data [Regier, 1984].

Hence researches in the 1960’s and 1970’s who had access only to systems like DSM-I were forced to rely on detailed description of symptom patterns – the most common approach was to employ a self reporting questionnaire or check list. For example General Health Questionnaire – GHQ [Golderg, 1970] and Hopkins symptoms check list. The St. Louis research criteria and the Research Diagnostic Criteria (RDC) were formulated in the 1970’s with corresponding interview schedules i.e. Research Diagnostic Interviews (RDI) and SADS [Kaplan, 1995]. Thus, a two stage methodology came into being and the use of  screening procedure was recommended as  the classical way of reducing the load of psychiatric interviewing in community research [Tarnopolsky, 1979]. Major efforts to develop standardised instruments to produce diagnostic interview protocols with acceptable inter-rater reliability resulted in the development of PSE-CATEGO Systems [Wing et al; 1974]. PSE-CATEGO system has an “Index of Definition” incorporated in a computer programme that is applied to the symptoms profile to make a diagnosis [Boyd, 1981]. In the UK also effort was undertaken to document the role of various social factors.

Thus the second generation studies introduced vigour in community surveys by giving attention to sampling, standardised techniques for analysis and also explored the role of social factors in mental illness. Their main limitation was that they did not generate rates  of  specific psychiatric disorders and obsured diagnostic variation [Weissmann, 1986].

4.0 Third generation studies:

With the advent of innovative diagnostic systems such as DSM-III [American Psychiatric Association (APA), 1980]. The diagnostic process was operationalised which led to the parallel development of a new generation of diagnostic instruments which revolutionised case identification in research [Mezzich, 1994].

These studies also used sophisticated techniques of sampling large community populations [Bland, 1988].

Development in other fields of psychiatric research led to advances in validity and reliability of diagnosis and also found strengthened evidence for biological factors in the aetiology of mental illness-hence the “Medical model” in this generation was reinstated [Weissmann, 1986].

Studies representative of this era are the Office of Population Censuses and Surveys (OPCS) surveys in Great Britain which used Schedules for Clinical Assessment in Neuropsychiatry (SCAN) [WHO, 1992] for confirmation. 14% of adults had a neurotic health problem and functional psychoses had a prevalence of 0.4% [Mason, 1986]. National Institute of Mental Health (NIMH)-Epidemiologic Catchment Area (ECA) programme which used Diagnostic Interview Schedule (DIS) [based on DSM-III) as a case identification instrument where 15.4% of the population above 18 years of age and over fulfilled criteria for atleast one  alcohol, drug abuse or other mental disorders. The rates for affective disorder was found to be 5.1% and anxiety disorder 7.3%. This one month prevalence of the ECA study was compared to other studies like European and Australian studies where prevalence varies from 9% to 16% [Regier, 1988].

6 month prevalence of NIMH (ECA) ranged from 15 to 23% which are similar to the rates of 16% found in Puerto Rico who used the Spanish version of the DIS [Hector, 1987].

5.0 General Practice Setting:

        Besides the community setting it is also now well established that within the general practice setting, 25-30% of patients have a psychiatric problem although their presenting complaints may often be physical in nature [Mann, 1997].

Surveys of the scope and range of psychiatric disorder in this setting began to appear in the 1960’s [Kessel, 1960; Shepherd, 1966]. Further surveys confirmed the overall magnitude of this problem [Goldberg, 1976].

6.0   Developing countries:

Compared to the west, in the developing countries, most fallacious was the belief that since people are not subjected to excessive demands by technological developments they are free from mental disorders [Leon, 1972].

Nothing was proved further from the truth when studies in these countries provided an impressive evidence of the overall magnitude of mental health problem. Neki (1973) – pointed out that even if one accepted a mental morbidity rate of 10 per 1000 population in the South-East Asian  region, they would have over 8 million psychiatric patients requiring active treatment.

The proportion of mental disorder In metropolitan urban zones of Latin American countries was found to between 17-18% of the population [Leon, 1972].

In rural Africa a community survey by Orley and Wing (1979) using the PSE found a prevalence rate of 27% in women and 17% in men.

                Research work in African population also indicated that mental disorders there may be as common as in the west [German, 1972]. In fact, in Ethiopia mental illness was more frequently seen in the hospital than infectious diseases [Giel, 1969].

                In 1975, the WHO began a multinational collaborative study involving seven developing countries including India to study the feasibility and effectiveness of offering community based mental health care.

                Their recorded frequency of psychiatric disorders was between 10.6% to 17.7% in primary health care setting [Harding et al; 1980].

7.0  Indian studies:

In the Indian setting, till about the first half of the century, no data was available regarding the prevalence of mental disorders [Srinivas Murthy, 1992].

The years 1962-1977 were marked by a large number of studies in the community and resulted in an increased awareness of the prevalence of such disorder in our country [Srinivas Murthy, 1992].

Most of these studies employed a 2-stage screening procedure but regrettably the majority of these studies utilised screening questionnaire that have not been adequately validated and used unstandardised clinical interviews for psychiatric assessment [Sen and Mari, 1987].

The area covered by these studies was also small and these studies have restricted to towns, rural areas and small cities [Chakorborty, 1990].

Some of these studies with prevalence rates in selected diagnostic categories are presented in the table.

Psychiatric morbidity (Rate per 1000 population)

                Psychiatric morbidity rate of Agra survey is significantly less than that observed in Lucknow and Vellore. Some difference could be due to methodological variation but the most plausible explanation for such a wide gap may be that the Agra survey was concerned only with severe psychiatric morbidity. The rate of mental retardation is low in Agra which appears to be gross underestimation in view of general consensus among the investigators of this county. The maximum difference exists in the rates of various types of psychoneurosis again being less in the Agra survey. Rates of schizophrenia and organic psychoses (not shown separately in the above table) were found to be similar.

Surya in 1964 surveyed an urban population in Pondicherry and found a prevalence rate of 9.5 per 1000. He also studied only major psychiatric illnesses.

The rural survey rates vary from 27 per 1000 [Ellnagger et al.] to 102.8 per 1000 [Nandi, 1975]. Sethi et al. (1972) reported a rate of 39 per 1000 in a survey of 500 families of a rural population.

Wide variation in the prevalence of neurosis was reported as shown in the table below.

One of the landmark studies in the Indian context was done on a rural population of South India (Kota) by Carstairs and Kapur. This was the first study in India where an attempt to use a standardised instrument was seen. They used the Indian psychiatric survey schedule (IPSS) which has been developed for psychiatric field work in Indian conditions. They put their case rate as 40% in women and 32% in men based on the presence of one or more somatic or psychological symptoms. They eschewed formal diagnostic categories. One outstanding feature of this study was comprehensive planning and skillful execution. But it came under severe criticism because if disability is determined only on the basis of symptoms elicited by the questionnaire very high figures of morbidity are obtained which do not convince clinicians of being an accurate reflection of the true state of affairs [Chakorborty, 1990].

Another survey done on a large scale was done in Calcutta with two schedules developed for structured interviews- a morbidity of nearly 109/1000 was found in which there was a high level of psychoneurosis [Chakorborty, 1990].

After the 70’s very few community surveys have been done in India. The most recent was done in rural Punjab using Bradford Somatic Inventory (VSI) and SRQ as screening questionnaire and diagnosis was done according to ICD-10 criteria. The BSI was thought to be culturally appropriate idiom of emotional distress. It was found that 46% of women and 15% of men suffered from anxiety and depressive disorder [Mumford, 1997].

Given the above epidemiological data on psychiatric morbidity, India’s commitment of mental health care forms an integral part of the total health programme and is endorsed within the objectives of NMHP.

8.0   Recent developments:

Presently standardised instrument like SADS, DIS, Clinical Interview Schedules (CIS) or PSE constitute the state of the art for establishing a psychiatric diagnosis [Burns, 1992].

The PSE which concentrates mainly on psychotic conditions has been used in major international studies such as the US-UK diagnostic project, the international Pilot study of schizophrenia and other studies of schizophrenia supported by WHO and NIMH. It is intended for use by skilled clinicians. In contrast DIS can be used by non clinicians to assess a large number of subjects.

But still comparable data between different nations won’t be possible until the approaches to case identification and diagnosis are similar.

Efforts are underway for the development of standardised instruments which are comparable with European, US and other diagnostic systems based on DIS and the PSE that will provide the basis for standardised generation of DSM-III and ICD-10 criteria. Initial drafts of this instrument called CIDI (Composite International Diagnostic Interview) are being reviewed by an international group of experts. Collaborative efforts between the Alcohol, Drug abuse and Mental health administration (ADAMHA) in the US and the WHO in Geneva are  underway to understand and empirically test the differences between European and American diagnostic approaches.

The developments under way offer promise that many of the previous difficulties will be resolved allowing for new direction in international psychiatric epidemiology.

MATERIALS AND METHODS

This is a community-based study.

1.0  Description of community  :   The community consists of approximately 5000 house-holds i.e. 27,368 individual living in an area covered by the Urban Health Centre, Mangalwar Peth, Pune.  The area covered by this centre is approximately 2 sq. Km.  It includes parts of Mangalwar Peth, New Mangalwar Peth and Somwar Peth.  It is located about 1 km. away from Sassoon General Hospitals and 1.5 km away from the Pune railway Station.

                        It is an Urban slum community which consists mostly of migrant population.

        There are 3 major sub-areas here, namely, plot no. 4 (predominantly Muslim population), Bhimnagar and Gadital (mainly Hindu population).  All individuals have access to safe water and electricity.  Most houses have a single room which is around 50-60 sq. feet. in area.  Although some families have a small bathroom (Mori) within the house, sanitation is mainly in the form of common municipal latrines and bathrooms.  Housing varies from temporary jute, plastic and wooden shelters to houses with mud or brick walls and tin roofs (pucca houses).  The area is served by 3 Municipal schools  and 1 Municipal General Hospital.

                Most houses have room with partitions in the form of curtains and woodsheets.  Some houses have lofts (malas) used for sleeping purposes and very few houses have more than one room.  Usually the flooring is of  plastered mud or stone (shahabadi stone).

Houses are poorly ventilated with just a single door and no windows.  Though the area is served by tarred roads, the interiors of the community do not have pucca roads. 

                This migrant population mostly comprises of people who are daily  wage  earners e.g. labourers.

2.0  Sample :

2.1 Sample selection :  

        In the initial stages of the study the sample was to be recruited from 4 sub-areas ( 3 were the above mentioned and 4th consisted of all remaining sub-areas) to ensure proportional representation. Subsequently each of these 4 areas were to be separately randomised.

        For this purpose, a list of the householders was sought from the Urban Health Centre.  As no comprehensive list was available there, the office of the District Collector, Pune Division and the Department of Food Grain Distribution were  consulted for the same.  On their recommendation ration shop owners in the area were consulted for the list.  These lists were either incomplete or had names which were mentioned more than once and hence were found to be inadequate for our purpose.  As fresh listing of the 5000 household within this community was beyond the scope of this dissertation, the bio-statistician was consulted who also visited the community for a field survey.

        Here it was realised that taking a small number of families from each of the bigger zones would not be representative of that area.

        In order to increase representativeness of the sample it was decided thereafter, that the sample would be selected from the Gadital area.  It was also found that using any specified method of randomisation in this area would be problematic due to the following reasons :

1.    A complete and reliable list of residents of this area was not available as mentioned above.

2.    The area is densely populated and houses are arranged in a very disorganised manner.

3.    Houses are located very close to each other and area is not divided into cluster by lanes or by-lanes.

As a result, it was advised by the statistician to resort to ‘quota’ sampling restriction which although is a non-random method of recruitment, ensures minimisation of biases and also ensures adequate representation of gender and age composition of the sample.

2.2 Sample characteristics :

Sample size: 150 families

The average family size was 5.6. For the purpose of the present study, based upon the observation of characteristics of the sample, socioeconomic status has operationally been categorised into lower and lower middle socioeconomic class. Most families had a kerosene stove, a 14 inch black and white television, a small portable radio and a bicycle (considered to be lower socio-economic class). Very few families also had a refrigerator, LPG connection, cemented houses with wooden doors (considered to be Lower middle socio-economic class). Almost every family had access to cable television. Individual houses were not numbered and the area number e.g. 232 Mangalwar Peth, suffices for the address.

2.3  Inclusion criteria:

                Individuals more than 18 years of age and of either sex currently residing in the community formed the study sample.

3.0  Tool :

SRQ-20 or Self Reporting Questionnaire :-

                SRQ is an instrument which was designed to screen for psychiatric disturbance in primary care settings. It was developed as a part of a collaborative study co-ordinated by WHO on strategies for extending mental health care.

                It was developed from four previously developed psychiatric research instruments (Harding et al. 1980) namely:

1.               PASSR (Climent and Plutchik, 1980).

2.               Post Graduate Institute Health Questionnaire (Verma and Wig, 1977).

3.               General Health Questionnaire (Goldberg, 1972).

4.               Present State Examination (Wing et al. 1974).

                The SRQ has now been extensively used in psychiatric epidemiology in many settings especially in the developing countries.

                Originally, the SRQ was conceived as a self completion questionnaire but it has been recommended that in countries where level of literacy is low, indirect application by health worker is more suitable (Mari, 1986).

                SRQ-20 is composed of 20 yes/no questions. Four on physical symptoms and 16 on psychoemotional disturbances (Appendix-II).

                The SRQ is designed to give only an estimate of illness. The patients psychiatric status has to be confirmed by a more extensive psychiatric interview.

                The questionnaire has been validated in many studies (Harding      et al., 1980 and Mari and Williams, 1986). It has already been used in people of different cultures.

                For the present study, a Marathi translation of the SRQ was used and back translation was done by a translator unacquainted with the original English version. Both the versions were compared and modified to achieve semantic equivalence of the items.

3.1  Augmentation of SRQ (Q. No 21-24) [Appendix III]:

In addition to SRQ-20 four questions on psychosis, alcohol, epilepsy and mental retardation were asked-purpose being to cover severe psychiatric morbidity as described in NMHP 1982 which can easily be done along side administration of SRQ-20.

                This Marathi translation was used for the purpose of the present study, namely validation of the screening tool to establish cut off scores for future use by the parent institute, if, so required.

4.0  Training of the investigator:

        Prior to data collection the investigator was trained to administer the Self Reporting Questionnaire (SRQ) by a qualified psychiatrists with specialised experience in psychiatric epidemiology within the community setting.

5.0  Data Collection:

                The epidemiological survey was conducted in two stages: a screening stage and a confirmation stage.

        Verbal consent for participation was taken in a culturally appropriate manner and confidentiality was ensured. Background sociodemographic information was documented in a specially designed proforma (Appendix-I).

5.1  Screening - stage:

                The screening stage consisted of a door to door survey in which the SRQ-20 was administered verbally to each person aged 18 years and over in the community setting.

                Administration of SRQ required around 7-10 minutes for each individual. Additionally, screening was also done on augmented SRQ for the presence of psychotic symptoms, mental retardation, epilepsy and alcohol dependence.

                Additional `30’ individuals were screened both on SRQ-20 and augmented SRQ from psychiatric Out Patient Department of the General Hospital, as it was though that the number of cases from the community alone would be inadequate for validation purpose considering the size of the sample and prevalence rates which would generate a very small number of  `cases’.  While retaining rater blindness to their `caseness’ status, the investigator administered the SRQ to these 30 individuals.

5.2 Confirmation stage:

                Assessment of caseness of the entire sample as well as the 30 individuals attending psychiatric out-patient clinic using clinical diagnosis was done by an independent qualified psychiatrist who was blind to the screen status of the individuals.

                Data was collected over a period of 6 weeks. Some degree of privacy was attempted while administering the questionnaire and during subsequent interviews. Many houses had to be visited on more than one occasion so as to complete the administration of the questionnaire to all the adult members of the household. The investigator had to visit families in the mornings and the evenings as well as on Sundays and Thursdays which are industrial holidays so as to accomplish target. A target of screening around 8-10 families per day was kept and validation was carried out later on the same day or a few days later, well within the time period where screening stage responses are considered valid (1 month).

6.0   Difficulties encountered:

1.               The physical setting of the survey created problems of privacy and at times the respondents were inhibited due to presence of other members of the family (for example, daughter-in-law responding in the presence of mother-in-law).

2.               At times, it was not possible to trace a family member even after the second or third visit to the same house. However such number was minimal in all 26 persons. The total sample coverage was 95%.

3.               Nature of interview was unsatisfactory in many cases (discussed later).

RESULTS

The results from the study are presented under the following Sections:

Section I  :       Demographic characteristics

Section II :       Screening stage

Section III       :       Confirmation stage

Section IV       :       Validation of the SRQ-20

Section I. Demographic characteristics (Chart I-IX)

·                   The total number of individuals in the 150 household were 844. Of these 495 were adults (above 18) year age group.

·                   Average family size was 5.6.

·                   Among the 495 adults male : female was found to be 1.02::1.

·                   Majority of the families were Hindus (68.7%) while others were Muslims (18%) and Buddhists (13.3%).

·                   Among the Hindus 26% were from the upper castes which consisted mainly of Brahmins and Marathas.

·                   There was slight preponderance  of nuclear families.

·                   98% of the families surveyed belonged to the lower socioeconomic status.

·                   62.3% of the surveyed population was literate.

·                   Of those individuals who are in the employable category, 59% were daily wage earners and 10.4% were unemployed.

·                   Age and gender distribution of the population showed that 1/3 of the total adult population was in the 21-30 years age group. Females outnumbered males in this age group as well as in 51-60 years age group.

GENDER DISTRIBUTION OF FAMILIES

(N=495)

DEMOGRAPHIC CHARACTERISTICS

CHART-I

RELIGIONWISE DISTRIBUTION OF FAMILIES

(N=150)

DEMOGRAPHIC CHARACTERISTICS

CHART – II

 CASTE WISE DISTRIBUTION OF FAMILIES

(N=150)

DEMOGRAPHIC CHARACTERISTICS

CHART – III

DISTRIBUTION OF FAMILY

BY STRUCTURE OF THE FAMILY

(N=150)

 

DEMOGRAPHIC CHARACTERISTICS

CHART-IV

                                                          DISTRIBUTION OF SAMPLE BY MARITAL STATUS

(N=495)

DEMOGRAPHIC CHARACTERISTICS

CHART – V

                                                                          DISTRIBUTION OF FAMILIES BY

                                                                               SOCIO-ECONOMIC STATUS   

DEMOGRAPHIC CHARACTERISTICS

CHART-VI

                                                         DISTRIBUTION OF SAMPLE BY EDUCATIONAL STATUS

                                                                                          (N=495) 

DEMOGRAPHIC CHARACTERISTICS

CHART – VII

DISTRIBUTION OF SAMPLES BY OCCUPATIONAL STATUS  (N=495)

DEMOGRAPHIC CHARACTERISTICS

CHART –VIII

AGE  AND GENDERWISE DISTRIBUTION OF SAMPLE (N=495)

Table 1.

DEMOGRAPHIC CHARACTERISTICS

CHART-IX

Section II. Screening stage:

                SRQ-20 was administered in Marathi to 469 adults representing 95% of the total adult population in the study frame.

339 adults had no scores on the SRQ i.e. 180 males and 159 females did no response on the  SRQ.

Table 2.  `YES’ responses on each item of SRQ-20 in the community.

Chart X: Scatter diagram showing itemwise scores on the

SRQ-20 for community.

Table 3 . `Yes’ responses on each item of SRQ-20 in clinic setting.

Chart XI: Scatter diagram showing itemwise scores on the

SRQ-20 for clinic.

Table 4 : Showing respondents with scores > 1 on SRQ-20 in the community setting.

Table 5. Showing respondents with scores > 1 on SRQ-20 in the clinic setting.

Mean score = 10.06; Standard deviation=4.5,

Coefficient of variation=0.44; Median =9.

Table 6. Showing score wise distribution on Augmented SRQ in community.

(Q.No. 21-24)

Table 7. Showing score wise distribution on Augmented SRQ in Clinic.

(Q. No. 21-24)

Section III : Confirmation stage :

Table 8. Showing scores of cases detected on SRQ screen morbidity in the community setting

Mean score =8.52; Standard deviation =2.4;

Coefficient of variation = 0.28; Median =8.

The scores of cases  in the clinic have been shown in section-II.

 [Table 5].

Pattern of psychiatric morbidity :

Confirmation was done by establishing `caseness’ of the entire screened sample by a qualified psychiatrist who was blind to the screen status of the individuals.

Table 9. Pattern of psychiatric morbidity [according to ICD-10 diagnosis] (number of cases) in community.

Total proportion of psychiatric morbidity in the population is 16.2%.

 Table 10. Showing pattern of psychiatric morbidity in the clinic.

 Table 11: Substance abuse in the community setting *.

 *  There were three cases of substance abuse (alcohol) in the clinic.

Chart XII : Substance abuse in the community setting .

Total = 43

% of males = [17.135]

Section IV. Relative Operating Characteristics (ROC curve) :

                ROC curve is obtained by plotting the sensitivity against the specificity for all possible cut-off points of the screening instrument because only a single pair of sensitivity and specificity for a particular score will not give a sufficient idea about the behaviour of the test which is under validation.

                Thus, a ROC curve can be described as summarising all the possible set of 2x2 decision matrices [Appendix IV] that result when the cut-off is varied from the largest to the  smallest possible value. For Plotting the ROC in this study 30 cases were taken from Psychiatric OPD, SGH and an additional 17 cases from the community.

Table 12. Sensitivity and specificity values for all possible cut-off points of the SRQ-20 (using 2x2 decision matrices).

Chart XIII: Relative Operating Characteristics (ROC) Curve.

DISCUSSION

For the purpose of convenience the results have been discussed sectionwise.

Section I: Demographic characteristic :-

Male to female ratio

Chart I. Shows that male to female ratio of the study  population was found to be 1.02:1. The gender ratio for the general population in India is 1.07:1 [Registrar General and Census Commissioner of India,, 1991].

Age and gender distribution (Table 1, Chart IX)

Age and gender distribution of the sample shows that 41% of the surveyed population was below 18 years. (40% of the population is below 15 years in developing countries compared to 15-20% in developed countries i.e. the population in developing countries is a young population). The population pyramid of the study sample in terms of age distribution is comparable to that of general population [Park, 1991].

Literacy (Chart VII)

Literacy in India according to 1991 census is 52.1 while the rate in Maharashtra is 72%. The present study found a literacy of 62.3%.

Occupation (Chart VIII)

Of those individuals in the employable category 59% were daily wage earners and 10.4% were unemployed.

Section II: Screening stage:-

The SRQ was administered to 95% of the total adults in  the population i.e. 469 adults. 26 adults did not undertake the screen as mentioned earlier in Section 6.0 (Materials and Methods). Out of 469 individuals 339 individuals i.e. 72% did not report any item affirmatively.

Items in SRQ essentially tap and are supposed to measure psychopathogical states such as anxiety, depression, somatization [WHO manual, 1994] amounting to respective illness.

Table 13 : Rank order of selected items in the community.

This table shows that three-quarters of the respondents reported affirmatively on item 6. However, this item does not refer exclusively to a mood or thought [WHO manual, 1994] and hence does not lend itself readily to a clinically meaningful interpretation. In order of frequency headache, poor appetite and bad sleep featured  high in rank order. Depressive symptoms are low in the rank order of frequency which probably reflects their substantive salience.

In the clinic setting (Table 3, Chart  XI) there was 100% response on item no.6. Somatic symptoms were also reported prominently, like headache (66.6%), poor appetite (66.6%) and poor sleep (80%). In contrast to community setting depressive symptoms here were frequent with 70% reporting that they have lost interest in things and more than half (52.3%) of the clinic sample reported suicidal ideation.

That somatic complaints were frequent here compares with clinically based findings that psychosomatic complaints are the ones mostly quoted by the patients [Reichenheim, 1991].

Reichenheim (1991) in a maternal mental health survey in Rio de Janerio found that anxiety symptoms were the most frequent ones. 78% of the mothers answer that they were nervous, tense or worried. Depressive symptoms were not so frequent.

Shepherd (1977) pointed to the undifferentiated nature of psychiatric morbidity as there is lack of clear separation of the various psychological and somatic dimensions. But it must be borne in mind that Shepherd et al. observations were made 20 years ago when ICD-8 and ICD-9 lacked operationalised criteria for diagnosis. The contemporary classifications namely ICD-10 DSM-IV provide well defined diagnostic criteria to differentiate various anxiety and mood disorders from somatoform disorders. SRQ as a tool was constructed with a purpose of estimating prevalence of this spectrum of psychiatric morbidity in the era of lack of precision of diagnostic criteria and hence its continued use a screening tool in the present time has given rise to a peculiar situation of sensitive but poorly specific separation of the pool of probable cases from known cases.

On this backdrop and in the absence of a new screening tool appropriate for psychiatric epidemiology in a community setting, an elaborate validation exercise has been undertaken to examine the utility of the same for future epidemiological research of the parent institute.

A detailed analyses of the data gathered on the screen has been discussed below.

The study sample comprises of two sub-samples (community and out patient department of Psychiatry i.e. clinic setting)  and data on both these sub-samples has been analysed separately so as to examine comparability of the results between both these samples.

The distribution of range of scores in community sample is 1-13 with no respondent scoring above 13. (Table 4). This is true even of respondents confirmed as cases (Table 8). Cases have aggregated on the higher side of the range of scores i.e. between 6-13 compared to proportion of respondents in the lower range of scores i.e. 1-5/6 (Table 8).

Likewise the clinic samples SRQ scores fall in the higher side of range between 6-20. 27 respondents fall in this range compared to 3 respondents in the range of 1-5 (Table 5).

Comparison of central tendencies such as mean, median, standard deviation and coefficient of variance using non-parametric tests namely    Mann-Whitney test showed no statistically significant difference between the pattern of scores of the two study samples.

Hence it may be concluded that subsequently computed validation indices of the clinic samples can be variably extrapolated to the community.

On the augmented SRQ in the community (Table 6) 43 subjects responded to alcohol abuse and 5 responded to item no. 21 (on psychotic symptoms). There were 3 subjects who responded to alcohol abuse in augmented SRQ in the clinic setting (Table 7).

Section III: Confirmation stage :-

The patients caseness status was confirmed by clinical interview using ICD-10 criteria (Table 9 and Table 10). All the 469 individuals from the community samples as well as the 30 individuals from the clinic samples were examined using clinical interview method. The status of caseness of the entire study sample has been presented in Table 14 (a) and 14 (b).

Table 14 (a) : Community setting (N=469).

Table 14(b) : Clinic setting (n=30).

                The pattern of morbidity of these cases has been discussed later.

Section IV: Validation of the SRQ-20 :-

                Validation of the tool for the specific setting has been undertaken so as to find its optimum use as screen for the purpose of this study. For the determination of caseness a psychiatric interview using  ICD-10 criteria was used.

                Out of an awareness that the sensitivity and specificity figures for a single cut-off score was probably not the most sophisticated was of presenting results, graphical way was found to be useful i.e. the ROC curve.

                Goldberg and Williams (1988) sum the advantages of  using the ROC analysis compared with giving a single value of sensitivity and specificity at one cut-off point.

-                    Assessment of the discriminating ability of the instrument across the total spectrum of morbidity.

-                    Comprehensive comparative assessment of the performance of two or more screening tests.

-                    Assessment of the effect of varying the threshold score.

The optimum cut-off point in this study was found at 6/7 where sensitivity was 85% and specificity was 83%.

Having established the cut-off on the screen using the data on the sub-population on the clinic, it is imperative to examine the predictive behaviour of the screening tool (post hoc) on the community sample.

Earlier in section II it has been discussed that the two sub-populations are comparable with  respect to their screening status and that there are no statistically significant differences in the distribution of SRQ scores in the cases of either sub-population.

                Using a cut-off of > 6, the community sample would have yielded 27 SRQ positive respondents.

Post hoc analysis:

Table15: Community sample (N=469).

                Of these 27 SRQ positive, 18 were cases and 9 were non-cases i.e. they were false positive on SRQ.

                Among the 442 SRQ negative pool there were three cases i.e. they were false negative on SRQ.

Table 16 : Showing estimated sensitivity and specificity post hoc.

Post hoc

Sensitivity 90% False positive –2%.

Specificity 98% False negative – 10% .

Table 17.                        Actual                     Estimated

The SRQ has already been tested for validity in a series of studies carried out between 1978 and 1993 as shown in the table below.

Table 18 : Sensitivity and Specificity figures of the SRQ-20.

        Present study                           6/7             85%                      83%

                The validation coefficients obtained in the present survey are intermediate between those reported by Harding et al. 1980 and Dhadphale et al. 1983. However, it is important to consider some methodological limitations in the studies described. Harding et al. used the PSE as the validating interview which has been developed primarily for psychiatric patients samples and may not be appropriate for community settings [Dohrenwend et al., 1978].

                Also Harding et al. 1980 applied the psychiatric interview to highly skewed sub-sample of respondents (90% of the high SRQ scores and only 10% of low scores). It is not clear whether the validity coefficients were subsequently adjusted so as to be representative of the screened sample as a whole [Diamond and Lilienfeld, 1962]  and adjustment which is crucial. Similarly it is not clear whether the coefficients reported by Dhadphale et al. 1983 were adjusted in like manner – a study which compared known patients with known normals and which yielded very high values for validity coefficients.

                As seen in the Table 18, sensitivity figures range from 62.9% to 90% whereas specificity figures range between 44% and 95.2%. The variability of the validation indices highlights the fact that a screening instrument needs to be validated in a variety of settings in different population. In other words, socio-cultural characteristics of the population may have an effect on the cut-off point e.g. in Sao Paulo a lower cut-off point in men gave the best results [Mari & Williams, 1985]. There is also evidence that individuals with little or no education are more likely to be false positive on the SRQ than those who have had more than eight years of schooling [Mari & Williams, 1986] and a simplification of the screening questionnaire in different settings (e.g. reducing the number of question) can be done if, so required.

                These points emphasis the recommendation [Harding et al., 1983] that the SRQ should be adequately validated for determination of cut-off point before being employed for large scale community surveys which has been done in this study.

Pattern of Psychiatric morbidity:

                Findings reported in this study are the proportion of individuals with mental morbidity in the surveyed population. No prevalence rates will be quoted owing to the small sample size and methodological limitations.  It is necessary to use standard method for confirmation which would permit diagnosis of entire range of psychiatric morbidity in a pre-determined uniform manner. In the absence of this prevalence rates i.e. definite estimate of presence or absence of a given category is not possible.

                The total proportion of psychiatric morbidity was 16.2%. One month prevalence estimates of psychiatric morbidity are usually found to be from 8% to 15% [Surtees et al., 1986].

                One month prevalence of mental disorders in the US were  determined from 18571 persons interviewed in all five sites that constitute NIMH-ECA programme and using DIS based on DSM-III criteria. It was found that 15.4% of the population 18 years of age and over fulfilled criteria for atleast one alcohol, drug abuse or other mental disorder during the period one month before interview. [Regier et al., 1988].

                These rates were compared to international epidemiological studies that use PSE for clinical assessment of subjects and quote one month prevalence rates. In European and Australian overall rates vary about 9% to 16% [Regier et al., 1988].

                17.13%  of the total male population was alcohol dependent. No female cases were identified. Alcohol dependence was maximum in males in the 41-50 yr. age group where 32.5% of males were  dependent. In an Indian study, in a urban community of Pondicherry alcohol dependence was a major social problem in the study area, affecting 7% of the adult males.

                In the OPCS survey in Great Britain which scanned over 18,000 addresses using CIS-R and SCAN for confirmation on individuals aged 16 to 64 years found that overall rate of alcohol dependence was found to be 4.7% and men were three times more likely than women to have alcohol dependence and dependence was most prevalent among young adults particularly males in 16-24 yr. age group. The prevalence rates in the OPCS study are remarkably similar to the rates found in the ECA study although excess of alcohol abuse in males is less marked than that found in the ECA study. [Mason et al., 1996].

                Even if we do not consider the proportion of cases that were underreported, alcohol dependence seems to be a major problem in this urban slum community.

                Schizophrenia was diagnosed in 2 cases (0.42%) in the surveyed population. Studies by Sethi (1967) and Verghesee (1973) reported a prevalence of 0.22% and 0.26% respectively.

                Verghese (1973) conducted a survey on 539 families in Vellore town using a validated mental health item sheet and 1965 ICD.

                Sethi (1967) interviewed each family member by questionnaire especially designed for the purpose as a screening tool and the diagnostic classification was done using DSM-II.

                In NIMH-ECA study one month rate of schizophrenic disorders was 0.7% with identical rates for both men and women.

                There were two cases of Bipolar affective disorder (0.42%) found in the study. In the NIMH-ECA study bipolar-I disorder had a relatively equal gender distribution with a combined male-female one month rate of 0.4%. The life time risk of bipolar disorder is less than 1%.

                There were 2(0.42%) cases of mental retardation. The rate of mental retardation estimated by the Expert Committee on National Mental Health Programme (1981) was 0.5-1%. The Lucknow study (Sethi) used Intelligence Quotient (IQ) tests and reported overall rate of retardation to be 1.2%.

                There were 6 cases of epilepsy (1.2%) found in the population surveyed. The rate for epilepsy in developing countries has been found to be 0.6-1% by the WHO studies [WHO Technical Report Series, 1978].

                In this study 11 cases (2.3%) were found in the Neurotic, stress related and Somatoform disorder category. These were mostly adjustment and anxiety disorders. There were 10 cases (2%) of depressive episode and dysthymia. Due to differences in case definition and diagnostic criteria rates of neurosis vary widely in national as well as international studies. Sethi found a rate of 2.71% in Lucknow and Verghese reported a rate of 4.76%.

                The rates were broadly similar to those found in the ECA study in which dysthymia was the most prevalent affective disorder at 3.3%. Male-Female one month rate of Bipolar affective disorder was 0.4%. Anxiety disorders in the ECA study had a rate of 7.3% and they were the most prevalent of the major groups of mental disorders. Somatization disorder was found in only 0.1% of the population there.

                The OPCS survey in Great Britain showed that 14% of adult had a neurotic health problem, mixed anxiety and depressive disorder being the commonest (7.7%) followed by generalised anxiety disorders (3.1%).

                That these disorders were proportionately more in women in the study keeping in with generalisations about the risk women carry for these type of disorders [Kessler, 1988].

Section V : Critical appraisal of the studies in community setting :-

                At the outset investigator of the present study would put on record with disheartening fact that the difficulties/limitations of studies in the community setting, which has direct implication on reliability and interpretation of data have been underreported in conventional epidemiological literature.

                A vital issue that came up during the administration of the SRQ in the community setting was the nature of the interview which was unsatisfactory due to many reasons.

                The physical setting of data collection had obvious limitations like lack of privacy, crowding around, resulting in inhibited responses or hurried answers.

                Although consent was not refused outright, the degree of co-operation obtained was variable – like the husband refusing to let his wife have a private dialogue or one member (usually the head) trying to represent all the other members of the family. A lady with one of her sons in jail gave vehement `No’ answers to all questions although there was enough circumstantial evidence for the presence of a disorder.

                Another important issue was the difference in conceptualisation of the question by the investigator and by the subject. Many answers appeared not to reflect the condition intended to be screened. For example question like always feeling tired, having stomach problems almost invariably had the effect of provoking a detailed account of physical illnesses in the past and questions like feeling that life is useless, being unhappy and feeling tense and worried were considered to be natural and accepted consequences of poverty and in no way reflecting an illness. The question about difficulty in decision  making had to be repeated many times over without a satisfactory answer.

                In the community setting, motivation for giving a `no’ answer was found to be more prominent then giving a `yes’ answer in contrast to primary health care setting.  Hence the vital issue concerning the low concept and criterion validity of the SRQ emphasised the importance of vigorous testing of the instrument in different socio-cultural settings and institution of modifications if required to carry across the same concept and purpose as intended.

                As stated earlier this study adopted a two-staged methodology to assess psychiatric morbidity in the community. But discussion of the results just presented will be incomplete if they are not highlighted in the light of methodological problems that plague psychiatric epidemiology.

                Kessel (1960) doubted the validity of international comparisons of psychiatric epidemiology.

                This is because it stands to reason that if the frequency of occurrence of an entity in two different situations is to be compared, the entity should be similarly defined  in the first place. This refers to the primary problem of case identification in psychiatric epidemiological studies. As each culture has its own criteria of reality, a symptom indicative of mental abnormality varies from one group to the other [Carstairs and Kapur, 1976].

                The second major problem is the different methods of classifying signs and symptoms into diagnostic groups and differences in diagnostic schemes in countries over time [Kramer, 1969].

                To deal with such problems various tests, scales and instruments have been devised [Chakorborty, 1990]. Objective questionnaires based on standard operational definitions of symptoms have been devised and laborious translation and retranslation of questions have been done in different languages and different settings [Carstairs and Kapur,1976].

                The SRQ has also been translated into a variety of languages to allow it to be used among people of different cultures [Kortmann and TenHorn, 1988].

                Although self-reports are economical, do not require extensive rater training and avoid the problem of observer bias [Myers, 1980], they have certain flaws :

1.               The number of questions limit the range of inquiry, so that some aspects of pathology will be missed if there are no questions to tap them. For example, the content validity of the SRQ with respect to covering all the mental disorders is fairly low. On the other hand the content validity related to covering neurotic disorders is high [WHO manual, 1994].

2.               The rigidity of the prescribed question, a factor deliberately introduced in order to increase reliability, prevents the kind of informal cross-examination which a clinician used while making his judgement about the patients state .

3.               When using a questionnaire the presence or absence of a symptom is based only on the patients response and thus the advantage of the clinicians experience gained over years of contact with psychiatric patients and of his ability to take account of the latter’s response set while making his decision is not available (which is why so many interviews were unsatisfactory in this study).

                This often makes a questionnaire lose in validity what it gains in reliability [Carstairs and Kapur, 1976].

                Another problem is that validation of the screening measures is absolutely dependant upon what Robins (1985) called erratic standards of clinical interview i.e. psychiatric assessment has to be assumed to be correct.

                Hence, a standardised procedure of examination like a semi-structured interview is widely recommended [Spitzer et al., 1975].

                But considering all these problems in perspective, even for validity results that are well below ideal, need not cause us to abandon the use of interviews tested. It is like comparing roentgenogram to a biopsy in a population survey. No one suggests abandoning roentgenograms because they clearly improve diagnosis over physical examination or history alone even though they do not improve it as much as a biopsy would. And psychiatric interviews in a large scale epidemiological survey is just about as impractical as a biopsy is in screening the general population.

                Hence we cannot afford to abandon interviews that allow us to make a diagnosis even though they do not always agree with results by psychiatrists [Robins, 1985].

                Since we lack a definitive lab test for psychiatric diagnosis it is simply a matter of whether or not one finds it advantageous to choose a single cutting point on that accumulation of characteristics used to define a disorder beyond which disorders are defined as present [Robins,  1985].

                Hence it is now generally agreed that the two staged screening procedure which utilises a validated screening technique followed by a standardised psychiatric interview, is an improvement in estimating the prevalence of psychiatric disorders  in the general population.

 CONCLUSION

                This is a community based study which used standard two-staged methodology with the primary aim of validating the screening tool.

                Two subsamples-469 individuals from the community and 30 from the clinic were subjected to assessment of their psychiatric status using SRQ for screening and Clinical interview.

                Validation of SRQ was carried out using method of plotting an ROC on the pooled data of cases from both the samples. Detailed analyses on the data obtained on the SRQ along with its augmentation and post hoc estimation of validity indices for the community has been carried and results have been discussed in the light of relevant literature.

                The investigator has concluded the entire exercise as an intensive learning opportunity in Psychiatric Epidemiological Research in the community setting and has suggested an optimum reporting of facts in the scientific literature in the true spirit of enquiry.

                                                                                   APPENDIX I

FAMILY NO:� � �                                        DOE: �� ��   ��

FAMILY NAME (head)______________________________________

RELIGION:                    CASTE:                   ADDRESS:

SOCIOECONOMIC STATUS:

FAMILY MEMBERS:                            FAMILY TYPE:

TOTAL                   ADULTS         CHILDREN

� �                             � �                            � �

                                                                                     APPENDIX –II

A copy of the English version of the Self Reporting Questionnaire is shown below:-

1.     Do you often have headache?                                Yes/No.

2.     Is your appetite poor       ?                                              Yes/No.

3.     Do you sleep badly?                                                        Yes/No.

4.     Are you easily frightened?                                              Yes/No.

5.     Do your hands shake?                                             Yes/No.

6.     Do you feel nervous, tense or worried?                          Yes/No.

7.     Is your digestion poor?                                            Yes/No.

8.     Do your have trouble thinking clearly?                          Yes/No.

9.     Do you feel unhappy?                                             Yes/No.

10.   Do you cry more often than usual?                         Yes/No.

11.   Do you find it difficult to enjoy your daily activities?            Yes/No.

12.   Do you find it difficult to make decisions?                     Yes/No.

13.   Is your daily work suffering?                                  Yes/No.

14.   Are you unable to play a useful part in life?                   Yes/No.

15.   Have you lost interest in things?                                     Yes/No.

16.   Do you feel that your are a worthless person?                       Yes/No.

17.   Has the thought of ending your life been on your mind?        Yes/No.

18.   Do you feel tired at all the time?                                    Yes/No.

19.   Do you have uncomfortable feelings in your stomach?          Yes/No.

20.   Are you easily tired?                                               Yes/No.

                                                                                  APPENDIX III

Augmented SRQ-24 :

21.         Do you ever hear voices which are not present or

which other people cannot hear?                                    Yes/No.

22.   Does any one in your house have low intelligence?               Yes/No.

23.   Do you get fits?                                                       Yes/No.

24.   Do you/Does anyone in the family drink alcohol?         Yes/No.

                                                                                    APPENDIX –IV

2x2 DECISION MATRIX :

Sensitivity = A/(A+C).

Specificity = D/(B+D).

                                                                          BIBLIOGRAPHY

1.               Bland R.C. [1988]: Investigation of the prevalence of psychiatric disorder. Acta Psychiatrica Scandinavica, 77 (suppl .338); 7-16.

2.               Boyd H, Weissman M.M. [1981]: Epidemiology of affective disorders. Re-examination and future directions. Archives of General psychiatry, 38;1039-1046.

3.               Burns B.J. Community Mental Health, Murty R., Burns B.J. Proceedings of the Indo-US symposium. First edition. NIMHANS publications, Banglore.

4.               Carstairs G.M. and Kapur R.L. [1976]: The great Universe of Kota : stress, change and mental disorders in an Indian village, London. The Hogarth Press.

5.               Carstairs G.M., Kapur R.L. and Kapur M. [1974]: Indian Psychiatric Survey Schedule (IPSS) Social Psychiatry, 9; 61-69.

6.               Carswell J. [1924]: Some Sociological considerations bearing upon the occurrence, prevention and treatment of mental disorders. Journal of Mental Science, 70;347-357.

7.               Chakraborty A. [1990]: Social stress and mental health. A social psychiatric field study in Calcutta. 1^st edition. Sage publications, New Delhi.

8.               Dhadpale M., Ellison R.H. and Griffin L. [1982] :The frequency of mental disorders among outpatients at a rural district hospital in Kenya. The Central African Journal of Medicine, 28(4); 85-89.

9.               Diamond E.L. and Lilienfeld A.M. [1962]: Effects of errors in classification and diagnosis in various types of epidemiological studies. American Journal of Public Health, 52;1137-44.

10.         Dohrenwend B.P., Dohrenwend B.S., Neugebauer R. and Hitzig R. [1982]: Mental illness in the United Sates: Psychiatry, Vol. 3. Epidemiological estimates. New York, Praeger.

11.         Dube K.C. [1970]: A study of prevalence and bio-social variables in mental illness in a rural and an urban community in U.P. India. Acta Psychiatrica Scandinavica, 46; 327.

12.         Ellnagar M. and Ro M.N. [1971]: Mental health in an Indian rural community. British Journal of Psychiatry ,119;499.

13.         German A. [1972]: Aspects of clinical psychiatry in sub-saharan Africa. British Journal of Psychiatry, 121;461-79.

14.         Giel R. and Van Luijk J.N. [1969]: Psychiatric morbidity in a small Ethiopian town. British Journal of Psychiatry ,115;149-162.

15.         Giel R. and Harding T.W. [1976]: Psychiatric priorities in developing countries. British Journal of Psychiatry ,128;513-22.

16.         Goldberg D. and Williams P. [1988]: A user’s guide to the General Health Questionnaire NFER-NELSON, Windsor, Great Britain.

17.         Goldberg D.P. [1972]: The detection of psychiatric illness by Questionnaire; Oxford University Press, London.

18.         Goldberg D.P. and Blackwell B. [1970]: Psychiatric illness in general practice: a detailed study using new methods of case identification. British Medical Journal, (ii); 439-443.

19.         Harding T., DeArongo, M.Ballazar J. et al. [1980]: Mental disorders in primary health care - A study of frequency and diagnosis in four developing countries; Psychological Medicine, 10;231-242.

20.         Harding T.W., Climent C.E., Diop M.B., Giel R., Ibrahim H.H.A., Murthy S.R., Suleiman M.A. and Wig N.N. [1983]: The WHO collaborative study on strategies for extending mental health care: II:xxx development of new research methods. American Journal of psychiatry, 140;1474-80.

21.         Hector R., Janino P., Bird Patrick et al. [1987]: The prevalence of specific psychiatric disorders in Puerto Rico. Archives of General Psychiatry, Vol. 44.

22.         Kaplan H.I., Sadock B.J. [1995]: Comprehensive Textbook of Psychiatry, Volume 2, Sixth edition, Williams and Wilkins, Baltimore.

23.         Kaplan H.I., Sadock B.J. and Grebb J.A. [1994]: Synopsis of Psychiatry, Seventh edition, Williams and Wilkins, Baltimore.

24.         Kessel R. [1960]: Psychiatric morbidity in London general practice. British Journal of Preventive and  Social Medicine, 129;61-67.

25.         Kessler L., Burns B., Shapiro S., Tischler G. et al. [1988]: Psychiatric diagnosis of medical service users-evidence from the epidemiological catchment area programme. American Journal of Public Health ,77;18-24.

26.         Kortmann F. and TenHorn S. [1988]: Comprehension and motivation in responses to a psychiatric screening instrument validity of the SRQ in Ethiopia. British Journal of Psychiatry, 153;93-101.

27.         Kramer M. [1969] : Cross rational study of the diagnosis of mental disorders - origin of the problem. American Journal of Psychiatry, 125(suppl. 1);1-11.

28.         Leighton D.C., Harding J.S., Madelin D.B., Hughes C.C. and Leighton A.H. [1963]: Psychiatric findings of the Sterling Country study. American Journal of Psychiatry, 129;1021.

29.         Leon C. [1972]: Psychiatry in Latin America. British Journal of Psychiatry ,121; 121-36.

30.         Mann N. et al. [1997]: Psychiatry and primary care - current opinion in psychiatry, 10;168-172.

31.         Mari J.J. and Williams P. [1986]: A validity study of a psychiatric screening questionnaire (SRQ-20) in primary care in the city of Sao Paolo. British Journal of Psychiatry, 148;23-26.

32.         Mason P. and Wilkinson G. [1996]: The prevalence of psychiatric morbidity: OCP a survey of psychiatric morbidity in Great Britain, 168;1-3.

33.         Mechanic D. [1992]: Murthy R., Burns B.J. (eds.). Community Mental Health - Proceedings of the Indo-US symposium. First edn. NIMHANS publications, Banglore.

34.         Mezzich J.E. [1989]: An empirical prototypical approach to the definition of psychiatric illness. British Journal of Psychiatry, 134[suppl. 4]; 42-46.

35.         Mumford D.B., Saeed K., Ahmed I., Shazia L. and Mubbashar M.H. [1997]: Stress and Psychiatric disorders in rural Punjab - a community survey. British Journal of Psychiatry ,170;473-478.

36.         Myers J.K. and Weissman M.M. [1980]: Use of a self report symptom scale to detect depression in a community sample. American Journal of Psychiatry, 137; 1081-1083.

37.         Nandi D.N. et al. [1980]: Mental morbidity and urban life –an epidemiological study. Indian Journal of Psychiatry, 22(4);324-330.

38.         Nandi D.N., Ajmany S., Ganguli H., Banerjee G., Boral G.C., Ghosh A. and Sarkar S. [1975]: Psychiatric disorders in a rural community in West Bengal - an epidemiological study. Indian Journal of Psychiatry, 17;87.

39.         National Institute of Mental Health, United States, 1985, Washington D.C.U.S. Government Printing Office [1985].

40.         National Mental Health Programme [1982] for Indian Director General of Health Services. Ministry of Health and Family Welfare, New Delhi, India.

41.         Neki J.S. [1973]: Psychiatry in South East Asia. British Journal of Psychiatry, 123;257-69.

42.         Newton J. [1962]: Preventing mental illness - prevalence and distribution of illness. Oxford University Press, London.

43.         Orley and Wing [1979]: Psychiatric disorders in two african villages. Archives of General Psychiatry, 36; 513-520.

44.         Park J.E. and Park K. [1991] :Park’s textbook of Preventive and Social Medicine, Thirteenth edition, Banarsidas Bhanot, Jabalpur.

45.         Passamanick et al. [1956] : A study of mental disease in an urban population. American Journal of Public Health, 47;923-929.

46.         Regier D.A., Boyd J.N., Burke J.D., Rai D.S., Myers J.K., Kramer M., Robins L.N., George L.K., Karno M.and Locke B.T. [1988]: One month prevalence of mental disorders in the United States. Archives of General Psychiatry,45;977-986.

47.         Regier D.A., Myers J.K., Kramer M., Robins L.N., Blazer DDD.G., Hough R.L., Eaton W.W. and Locke B.T. [1984]: NIMH Epidemiologic Catchment Area (ECA) programme- historical context, major objectives and study population characteristics. Archives of General Psychiatry, 41;934-941.

48.         Reichenheim M.E. and Harpham T. [1991]: Maternal mental health in a squatter settlement in Rio de Janeiro. British Journal of Psychiatry, 159;683-690.

49.         Robins L.N. [1985]: Epidemiology - reflections on testing the validity of psychiatric interviews. Archives of General Psychiatry, 42;918-924.

50.         Salleh M.R. [1990]: Psychiatric morbidity in schizophrenic relatives – use of Self Reporting Questionnaire (SRQ). Singapore Medical Journal, 3195; 457-462.

51.         Satorius N. and Harding T. [1983]: The WHO collaborative study on strategies for extending mental health care- I- The genesis of the study. American Journal of Psychiatry, 140;1491-92.

52.         Sen B., and Williams P. [1987]: The extent and nature of depressive phenomenon in Primary Health Care-a study in Calcutta, India. British Journal of Psychiatry, 151;486-493.

53.         Sen B., Wilkinson G. and Mari J.J. [1987] :Psychiatric morbidity in Primary Health Care, a two stage screening procedure in developing countries-choice of instruments and cost effectiveness. British Journal of Psychiatry, 151;33-38.

54.         Sethi B.B., Gupta S.C. and Kumar R. [1967]: A psychiatric study of 300 urban families. Indian Journal of Psychiatry, 9;280-302.

55.         Sethi B.B., Gupta S.C., Mahendra R.K. and Promila K. [1974]: Mental Health and Urban life - a study of 850 families. British Journal of Psychiatry, 124;293.

56.         Sethi B.B., Gupta S.C. and Rajkumar  P. [1972]: Psychiatric survey of 500 rural families. Indian Journal of psychiatry, 14;183.

57.         Sheperd M. [1977]: Beyond the layman’s madness- The extent of Mental disease In: Developments in Psychiatric Research. J.M. Tanner (eds), London, Hodder and Stoughton.

58.         Shepherd M., Cooper B., Brown A. and Kalton G. [1966]: Psychiatric illness in General Practice, Oxford, Oxford University Press.

59.         Spitzer P.G. [1987]: Psychiatric disorders in the community and the GHQ. British Journal of Psychiatry, 150;828-833.

60.         Srinivasamurthy R. [1992]: Community Mental Health in India xxx Murthy R. Burns B.J. (eds.). Proceedings of the Indo-US Symposium, Community Mental Health, First edition, NIMHANS, Bangalore.

61.         Srole L., Langer T.S., Michael S.T. et al. [1962] :Mental Health in the Metropolis - The Midtown Manhattan study, New York, McGraw-Hill Book Co.

62.         Surtees P.G., Sashidharan S.P. and Dean C. [1986]: Affective disorder amongst women in the general population- a logitudinal study. British Journal of Psychiatry, 148;176-186.

63.         Surya N.C. [1964]: Mental morbidity in Pondicherry Transaction 4, All India Institute of Mental Health, Bangalore.

64.         Tamopolsky et al. [1979]: Validity and uses of a screening questionnaire (GHQ) in the community. British Journal of Psychiatry, 134;508-15.

65.         Thacore V.R., Gupta S.C. and Suraiya M. [1974]: Psychiatric morbidity in a North Indian Community. British Journal of Psychiatry, 126;361-369.

66.         Varquez-Barquero J.L., Diez Manrique J.F., Pena C., Quintanal R.G. and Lopez M.L. [1978]: Two stage design in a community survey. British Journal of Psychiatry, 149;88-97.

67.         Verghese A. and Beig A. [1979]: Neurosis in Vellore Town-an epidemiological study. Indian Journal of Psychiatry, 16;1.

68.         Verghese et al. [1973]: A social psychiatric study of a representative group of families in Vellore town. Indian Journal of  Medical Research, 61;608-620.

69.         Weissmann et al. [1978]: Psychiatric disturbance in a US urban community 1976-1978. American Journal of Psychiatry, 135;439-462.

70.         Weissmann M.M. [1986]: In- Psychiatry Cavenar (ed) Vol. 3, Epidemiology - purpose and historical overview. Basic Book Inc., Publishers. New York.

71.         WHO [1965]: International Statistical Classification of Diseases, Injuries , and Causes of Death. 8^th revision, Geneva, WHO.

72.         WHO [1975]: Organisation of  mental health services in developing countries. Technical Report Series No. 564, WHO, Geneva.

73.         WHO [1978] : International Classification of Diseases (9^th rev.), WHO, Geneva.

74.         WHO [1978]: The application of advances in neurosciences for neurological disorders. Technical Report Series No. 629, WHO, Geneva.

75.         WHO [1993]: ICD-10. The ICD-10 Classification of Mental and Behavorial Disorders: Clinical description and Diagnostic guidelines,  WHO, Geneva.

76.         WHO[1994]: A user’s guide to the Self Reporting Questionnaire (SRQ), Division of Mental Health, WHO, Geneva.

77.         Wing J.K., Cooper J.E. and Sartorius N. [1974]: The measurement and classification of psychiatric illness, Cambridge. Cambridge University Press.

S.

Please return again frequently as we have some rather BIG PLANS ahead for our web site. We intend to be so much more than just a peer reviewed online journal of  medicine. Once completed our web site will contains viable articles, expert opinions, letters, and extensively researched reports. All articles may be submitted by e-mail as well as by post and will be evaluated accordingly. We thank you for your patience.

Home | Dr. Ramesh Kaul | Contact Us | Submit Content
 

Thorax.us makes absolutely no representation as to the content, accuracy, or completeness of the information contained on this web site or the linked web sites and does not necessarily endorse any of the information contained therein. All readers and/or patients should verify all of the information and data before ever administering any type of drug, therapy or treatment discussed herein. Neither the web designers, editors nor the publisher accepts any responsibility for the accuracy of the information or the consequences from the use or misuse of the information contained herein. Information on this web site is provided for informational purposes only and is not a substitute for professional medical advice. You should not use the information on this web site for diagnosing or treating a medical or health condition. You should carefully read all product packaging. If you have or suspect you have a medical problem, promptly contact your professional healthcare provider. Statements and information regarding supplements have not been evaluated or approved by the Food and Drug Administration. Please consult your healthcare provider before beginning any course of supplementation or treatment.
 

Copyright © 2005 Thorax.us All Rights Reserved